Convergence speed improvement for a variable step-size plain gradient algorithm by using variable notch bandwidth technique

2004 ◽  
Author(s):  
R. Punchalard ◽  
W. Lertvasana ◽  
P. Chumchu
Keyword(s):  
Convergence Speed ◽  
Gradient Algorithm ◽  
Variable Step Size ◽  
Step Size ◽  
Variable Step

Active control for vehicle interior noise using the improved iterative variable step-size and variable tap-length LMS algorithms

2019 ◽  
Vol 67 (6) ◽  
pp. 405-414 ◽  
Author(s):  
Ningning Liu ◽  
Yuedong Sun ◽  
Yansong Wang ◽  
Hui Guo ◽  
Bin Gao ◽  
...  
Keyword(s):  
Steady State ◽  
Noise Control ◽  
Convergence Speed ◽  
Interior Noise ◽  
Lms Algorithm ◽  
Variable Step Size ◽  
Step Size ◽  
Vehicle Interior ◽  
Vehicle Interior Noise ◽  
Variable Step

Active noise control (ANC) is used to reduce undesirable noise, particularly at low frequencies. There are many algorithms based on the least mean square (LMS) algorithm, such as the filtered-x LMS (FxLMS) algorithm, which have been widely used for ANC systems. However, the LMS algorithm cannot balance convergence speed and steady-state error due to the fixed step size and tap length. Accordingly, in this article, two improved LMS algorithms, namely, the iterative variable step-size LMS (IVS-LMS) and the variable tap-length LMS (VT-LMS), are proposed for active vehicle interior noise control. The interior noises of a sample vehicle are measured and thereby their frequency characteristics. Results show that the sound energy of noise is concentrated within a low-frequency range below 1000 Hz. The classical LMS, IVS-LMS and VT-LMS algorithms are applied to the measured noise signals. Results further suggest that the IVS-LMS and VT-LMS algorithms can better improve algorithmic performance for convergence speed and steady-state error compared with the classical LMS. The proposed algorithms could potentially be incorporated into other LMS-based algorithms (like the FxLMS) used in ANC systems for improving the ride comfort of a vehicle.

A new adaptive variable step size natural gradient BSS algorithm

2021 ◽  
pp. 1-12
Author(s):  
Junqing Ji ◽  
Xiaojia Kong ◽  
Yajing Zhang ◽  
Tongle Xu ◽  
Jing Zhang
Keyword(s):  
Signal To Noise Ratio ◽  
Wavelet Coefficient ◽  
Gradient Algorithm ◽  
Variable Step Size ◽  
Natural Gradient ◽  
Threshold Method ◽  
Step Size ◽  
Morphological Component Analysis ◽  
Variable Step ◽  
Separation Degree

The traditional blind source separation (BSS) algorithm is mainly used to deal with signal separation under the noiseless model, but it does not apply to data with the low signal to noise ratio (SNR). To solve the problem, an adaptive variable step size natural gradient BSS algorithm based on an improved wavelet threshold is proposed in this paper. Firstly, an improved wavelet threshold method is used to reduce the noise of the signal. Secondly, the wavelet coefficient layer with obvious periodicity is denoised using a morphological component analysis (MCA) algorithm, and the processed wavelet coefficients are recombined to obtain the ideal model. Thirdly, the recombined signal is pre-whitened, and a new separation matrix update formula of natural gradient algorithm is constructed by defining a new separation degree estimation function. Finally, the adaptive variable step size natural gradient blind source algorithm is used to separate the noise reduction signal. The results show that the algorithm can not only adaptively adjust the step size according to different signals, but also improve the convergence speed, stability and separation accuracy.

STOCHASTIC INFORMATION GRADIENT ALGORITHM WITH GENERALIZED GAUSSIAN DISTRIBUTION MODEL

2012 ◽  
Vol 21 (01) ◽  
pp. 1250006 ◽  
Author(s):  
BADONG CHEN ◽  
JOSE C. PRINCIPE ◽  
JINCHUN HU ◽  
YU ZHU
Keyword(s):  
Gradient Algorithm ◽  
Distribution Model ◽  
Variable Step Size ◽  
Step Size ◽  
Dispersion Parameters ◽  
Gaussian Density ◽  
Variable Step ◽  
Generalized Gaussian Density ◽  
Automatic Switching ◽  
Gaussian Distribution Model

This paper presents a parameterized version of the stochastic information gradient (SIG) algorithm, in which the error distribution is modeled by generalized Gaussian density (GGD), with location, shape, and dispersion parameters. Compared with the kernel-based SIG (SIG-Kernel) algorithm, the GGD-based SIG (SIG-GGD) algorithm does not involve kernel width selection. If the error is zero-mean, the SIG-GGD algorithm will become the least mean p-power (LMP) algorithm with adaptive order and variable step-size. Due to its well matched density estimation and automatic switching capability, the proposed algorithm is favorably in line with existing algorithms.

scholarly journals Hybrid Filtered-x Adaptive Vibration Control with Internal Feedback and Online Identification

2018 ◽  
Vol 2018 ◽  
pp. 1-15 ◽  
Author(s):  
Lihua Yang ◽  
Shuyong Liu ◽  
Haipeng Zhang ◽  
Haiping Wu ◽  
Haifeng Li ◽  
...  
Keyword(s):  
Vibration Isolation ◽  
Mechanical Vibration ◽  
Convergence Speed ◽  
Variable Step Size ◽  
Control Effect ◽  
Step Size ◽  
Isolation System ◽  
Internal Feedback ◽  
Online Identification ◽  
Variable Step

Active control is an effective way to suppress low-frequency mechanical vibration. However, with applications to submarine equipment, there are still some shortcomings due to vibration coupling and multifrequency complex excitation. In this paper, a novel hybrid improved adaptive control strategy, feedback and online identification filtered-x LMS, namely, FOFxlms, is proposed, which introduces the residual errors to correct variable step-size, uses the estimated primary path to improve online identification, and applies internal feedback to compensate for the feedforward control. Then the FOFxlms algorithm is applied to a double-layer vibration isolation system of submarine rotating equipment, and the simulation results show that the normalized variable step-size with residual error can effectively improve convergence speed, the internal feedback can efficaciously compensate for steady-state control accuracy, and the online identification can dynamically identify the time-varying characteristics of the secondary path. The vibration reduction efficiency of Fxlms, FFxlms, and FOFxlms increases for the fundamental frequency vibration; the control effect and convergence speed are also enhanced in turn.

Sign in / Sign up

Export Citation Format

Share Document